The objective of this project is to complete an ongoing two arm (n = 126/arm) randomized, double-blind, placebo-controlled phase I I/I11 clinical trial in men who are at high risk for prostate cancer recurrence after a radical prostatectomy with administration of soy protein isolate or a casein-based placebo for up to two years. The dose is 20 g protein/day (-22-26 mg/day genistein and -38-42 mg/day total isoflavones). Preliminary results indicate that this treatment is extremely safe Serum PSA is being measured with an ultra- sensitive assay (which allows detection of PSA failure one year earlier than conventional assays) every 2 months in year 1 and every 3 months in year 2. We will accrue a total of 284 men (assuming a 13% drop-out rate based on preliminary results) from at least eight clinical sites around the USA, two of which also have a sub-site at a VA Medical Center. Comparing the treatment and placebo groups, the primary end-point is the rate of PSA failure as surrogate for prostate cancer recurrence. Based on literature data and the characteristics of already enrolled men, we expect a 30% rate of PSA failure in two years in this group of men who must fulfill at least one of the following eligibility criteria: pre-operative PSA >20 ng/ml, Gleason sum of > 7, positive surgical margins, extracapsular extension, seminal vesicle invasion, and/or lymph node (micro) metastases. The sample size of 126/arm will allow us to detect a 40% reduction in the rate of PSA failure from 30% to 18% with 80% power at a significance level of 0.05. We will also determine whether the soy intervention affects time-to-PSA failure, serum cholesterol, and serum isoflavone levels. In addition, effects will be determined on testosterone, estradiol, and SHBG (steroid hormone axis), and T3 and T4 (thyroid activity). This study will yield information about ability of soy protein containing isoflavonesto prevent recurrence of prostate cancer after radical prostatectomy. The study will also provide information about of the preventive potential of soy protein in men that have not been diagnosed with prostate cancer, because there is a high prevalence (>50%) in these men of undetected histologic prostate carcinomas.